The purpose of this paper is to describe the place of twin studies in etiological research on somatic diseases, using asthma as an example. Twin studies provide answers to the relative importance of genes and environment in the development of disease, but have been criticised for systematic biases especially linked to unusual conditions in twin pregnancies. For asthma, classical twin studies show that genes explain the largest part of the interindividual variability, and that common family environment does not explain similarity in asthma in siblings. The assumptions underlying classical twin studies are discussed. A few other twin designs are mentioned, and we conclude that twin studies still play an important role in understanding causes of somatic diseases.